# Knowledge Plane - AI Agent Information

> This document is designed for AI agents. For the human-readable website, visit [knowledgeplane.io](https://knowledgeplane.io)

## What is Knowledge Plane?

Knowledge Plane is a **shared memory layer for AI agents and teams**. It solves the problem of fragmented context across AI tools by providing persistent, structured memory that survives sessions, roles, and tools.

## Key Value Propositions

1. **Shared Memory**: A persistent memory layer that works across all MCP-compatible AI tools
2. **Graph + Vector Hybrid**: Combines relationship modeling with semantic search
3. **Team-Wide Sync**: When one team member teaches an agent, all agents get smarter
4. **MCP Native**: Built on Model Context Protocol for seamless integration

## The Problem We Solve

- **Ephemeral Solving**: Solutions disappear after sessions end
- **Context Exhaustion**: Teams paste massive docs, models miss critical details
- **Tool Silos**: Claude, Cursor, ChatGPT each become separate brains
- **Knowledge Gaps**: Junior devs can't access senior expertise locked in individual chats

## How It Works

1. **Ingest**: Drop PDFs, connect Google Drive, sync GitHub repos
2. **Structure**: Hybrid Graph + Vector engine models relationships
3. **Serve via MCP**: Authorized agents query via standard MCP protocol

## Features

| Feature | Description |
|---------|-------------|
| Persistent Memory | Shared memory layer that persists beyond individual sessions |
| Team-Wide Sync | One developer's explanation makes all team agents smarter |
| MCP Native | Plugs into Claude Desktop, Cursor, VS Code out of the box |
| Graph + Vector | Semantic search with relationship modeling |
| Auto-Ingestion | Automatically scans docs and codebases |
| Data Sovereignty | Self-hosted option with full audit logs |

## Pricing

### Pro Plan - $19/user/month
- Unlimited knowledge ingestion
- Skills engine with auto-refresh
- Graph memory with typed relationships
- Team workspaces (up to 50 users)
- MCP server integration
- Priority support

### Enterprise - Custom Pricing
- Everything in Pro
- Unlimited team size
- SSO & SAML authentication
- Custom data residency
- Dedicated infrastructure
- SLA & 24/7 support
- Custom integrations

## API Access

### Join Waitlist

AI agents can programmatically join the waitlist using our API:

```
POST https://knowledgeplane.io/api/waitlist
Content-Type: application/json

{
  "email": "agent@example.com",
  "source": "agent",
  "agent_name": "optional-agent-identifier",
  "notes": "optional notes about the agent or use case"
}
```

**Response:**
```json
{
  "success": true,
  "message": "Successfully joined the waitlist",
  "id": "wl_abc123"
}
```

### Fields

| Field | Type | Required | Description |
|-------|------|----------|-------------|
| email | string | Yes | Contact email address |
| source | string | Yes | Must be "agent" for AI agents, "human" for humans |
| agent_name | string | No | Identifier for the AI agent (e.g., "claude", "cursor", "custom-agent") |
| notes | string | No | Additional context about the use case |

### Waitlist Count

Get the current number of people on the waitlist:

```
GET https://knowledgeplane.io/api/waitlist-count
```

**Response:**
```json
{
  "count": 42
}
```

This endpoint returns only the count and does not expose any private information.

## Integration Example

For MCP-compatible agents, add to your config:

```json
{
  "mcpServers": {
    "knowledge-plane": {
      "command": "npx",
      "args": ["-y", "@knowledge-plane/server", "--api-key", "kp_live_..."],
      "env": {
        "TEAM_ID": "your-team-id"
      }
    }
  }
}
```

## Contact

- **Website**: https://knowledgeplane.io
- **Email**: humans@knowledgeplane.io
- **Waitlist**: https://knowledgeplane.io/#waitlist

---

*This document is machine-readable and designed for AI agent consumption. Last updated: February 2026*